本文選題：信息融合 + 態(tài)勢估計(jì)�。� 參考：《杭州電子科技大學(xué)》2017年碩士論文

【摘要】：現(xiàn)代信息戰(zhàn)爭中呈現(xiàn)出多樣化的作戰(zhàn)方式、多元化的作戰(zhàn)對象以及復(fù)雜多變的作戰(zhàn)環(huán)境等特點(diǎn)。這就要求現(xiàn)代C4ISR系統(tǒng)能快速準(zhǔn)確的融合大量多元異構(gòu)的戰(zhàn)場數(shù)據(jù)信息,為形成作戰(zhàn)決策提供證據(jù)支持。因此,戰(zhàn)場中的態(tài)勢估計(jì)研究成為多傳感器信息融合領(lǐng)域的一個(gè)重要課題。在實(shí)際應(yīng)用中,如何對多傳感器提供的大量數(shù)據(jù)信息作出相關(guān)性分析;如何發(fā)現(xiàn)并利用少量但重要的沖突信息;如何對融合后態(tài)勢結(jié)果做出一致性評價(jià);如何解決融合時(shí)大量數(shù)據(jù)信息會(huì)出現(xiàn)的冗余問題,是目前態(tài)勢估計(jì)研究亟需解決的若干難點(diǎn)和關(guān)鍵問題。針對上述問題,本文針對態(tài)勢估計(jì)中沖突數(shù)據(jù)信息,以提高融合結(jié)果的魯棒性及可信度為目標(biāo)開展研究。主要的研究內(nèi)容如下:首先,簡述了本課題研究的背景和意義,以及對態(tài)勢估計(jì)及其一致性的國內(nèi)外研究現(xiàn)狀進(jìn)行綜述。然后詳細(xì)討論了在態(tài)勢估計(jì)過程中導(dǎo)致不確定性信息問題出現(xiàn)的原因,并介紹了處理態(tài)勢估計(jì)中不確定性問題的一些典型方法,為下文對態(tài)勢估計(jì)問題的研究奠定基礎(chǔ)。其次,針對基于D-S證據(jù)理論的態(tài)勢估計(jì)方法在處理多源沖突數(shù)據(jù)時(shí)融合效果不佳的問題,第3章提出了一種基于沖突數(shù)據(jù)聚類的態(tài)勢估計(jì)方法。首先結(jié)合Jousselme距離和傳統(tǒng)沖突系數(shù)構(gòu)建一種新的沖突證據(jù)表征方式;然后利用迭代自組織數(shù)據(jù)聚類方法對數(shù)據(jù)進(jìn)行聚類;最后對不同聚類簇的證據(jù)采用D-S理論融合得到態(tài)勢結(jié)果,同時(shí)構(gòu)建距離準(zhǔn)則函數(shù)評價(jià)態(tài)勢結(jié)果的一致性。仿真結(jié)果表明:與傳統(tǒng)態(tài)勢估計(jì)方法相比,本文所提方法在融合多源沖突數(shù)據(jù)時(shí)能夠得到可信度較高的態(tài)勢估計(jì)結(jié)果。再次,異類傳感器產(chǎn)生大量冗余、沖突的信息,導(dǎo)致常規(guī)態(tài)勢估計(jì)方法性能下降的問題,第4章提出基于異類傳感器信息自適應(yīng)融合的魯棒態(tài)勢估計(jì)方法。首先將大量傳感器分為兩組——全天候和輔助型傳感器;然后構(gòu)造兩級融合結(jié)構(gòu),并基于Jousselme距離評估融合結(jié)果的一致性;最后在此基礎(chǔ)上自適應(yīng)地通過兩級信息融合得到態(tài)勢估計(jì)結(jié)果。仿真結(jié)果表明,本文所提方法能夠在兼顧運(yùn)算效率的基礎(chǔ)上提高態(tài)勢估計(jì)結(jié)果的魯棒性。最后,對本文所研究的問題進(jìn)行了總結(jié)與展望。
[Abstract]:In modern information war, there are many characteristics, such as diversified combat mode, diversified combat object and complex and changeable combat environment. This requires that modern C4ISR systems can quickly and accurately integrate a large number of heterogeneous battlefield data information and provide evidence support for the formation of operational decisions. Therefore, the research of situation estimation in battlefield becomes an important subject in the field of multi-sensor information fusion. In the practical application, how to make the correlation analysis to the massive data information provided by the multi-sensor, how to find and utilize the small amount of but important conflict information, how to make the consistency evaluation to the result of the fusion situation; How to solve the redundant problem caused by a large amount of data information in fusion is a difficult and key problem that needs to be solved in the research of situation estimation at present. In order to improve the robustness and reliability of fusion results, this paper aims to improve the robustness and credibility of fusion results by focusing on the conflict data information in situation estimation. The main research contents are as follows: firstly, the background and significance of this research are briefly introduced, and the current research situation of situation estimation and its consistency is summarized at home and abroad. Then, the causes of uncertain information problems in the process of situation estimation are discussed in detail, and some typical methods to deal with the uncertainty problems in situation estimation are introduced, which will lay a foundation for the research of situation estimation problems below. Secondly, aiming at the problem that the situation estimation method based on D-S evidence theory is not effective in dealing with multi-source conflict data, chapter 3 proposes a situation estimation method based on conflict data clustering. Firstly, a new representation of conflict evidence is constructed by combining Jousselme distance and traditional conflict coefficient. Then, iterative self-organizing data clustering method is used to cluster the data. Finally, D-S theory is used to fuse the evidence of different clusters to obtain the results. At the same time, the distance criterion function is constructed to evaluate the consistency of situation results. The simulation results show that compared with the traditional situation estimation method, the proposed method can obtain a more reliable situation estimation result when the multi-source conflict data are fused. Thirdly, heterogeneous sensors produce a lot of redundant and conflicting information, which leads to the performance degradation of conventional situation estimation methods. In chapter 4, a robust situation estimation method based on adaptive fusion of heterogeneous sensor information is proposed. Firstly, a large number of sensors are divided into two groups-all-weather and auxiliary sensors, and then two levels of fusion structure are constructed, and the consistency of fusion results is evaluated based on Jousselme distance. Finally, the situation estimation results are obtained by adaptive two-level information fusion. The simulation results show that the proposed method can improve the robustness of the situation estimation results on the basis of taking into account the computational efficiency. Finally, the problems studied in this paper are summarized and prospected.
【學(xué)位授予單位】：杭州電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：E86;TP212

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 趙宗貴;許陽;;信息融合動(dòng)態(tài)與發(fā)展趨勢[J];指揮信息系統(tǒng)與技術(shù);2014年06期

2 孫強(qiáng);岳繼光;;基于不確定性的故障預(yù)測方法綜述[J];控制與決策;2014年05期

3 曾清;施慧杰;杜陽華;;聯(lián)合作戰(zhàn)戰(zhàn)場態(tài)勢一致性評估[J];指揮控制與仿真;2014年01期

4 李峗;劉鋼;老松楊;;A genetic algorithm for community detection in complex networks[J];Journal of Central South University;2013年05期

5 李君靈;王裕;趙宗貴;;多類差異信息柔性融合概念與內(nèi)涵[J];指揮信息系統(tǒng)與技術(shù);2013年02期

6 王連鋒;宋建社;朱昱;曹繼平;;基于模糊聚類分析的證據(jù)組合[J];系統(tǒng)工程與電子技術(shù);2013年01期

7 羅佳;黃璽瑛;高會(huì)波;;基于多重信息單元的戰(zhàn)場態(tài)勢要素構(gòu)建研究[J];國防科技;2012年03期

8 趙宗貴;李君靈;王珂;;Context概念與內(nèi)涵及其在信息融合中的應(yīng)用[J];指揮信息系統(tǒng)與技術(shù);2012年02期

9 裴曉黎;;面向服務(wù)的海軍作戰(zhàn)與仿真一體化構(gòu)建技術(shù)[J];艦船電子工程;2011年12期

10 胡麗芳;關(guān)欣;鄧勇;何友;;一種三角模糊數(shù)型多屬性決策方法[J];控制與決策;2011年12期

相關(guān)碩士學(xué)位論文 前2條

1 黃玉奇;基于貝葉斯網(wǎng)絡(luò)和本體的態(tài)勢估計(jì)方法[D];杭州電子科技大學(xué);2012年

2 胡小佳;態(tài)勢估計(jì)中的不確定性推理方法研究[D];國防科學(xué)技術(shù)大學(xué);2007年

，

本文編號：1979678